init datamate

runtime/ops/mapper/garble_characters_cleaner/__init__.py

@@ -0,0 +1,6 @@
+# -*- coding: utf-8 -*-
+
+from datamate.core.base_op import OPERATORS
+
+OPERATORS.register_module(module_name='GrableCharactersCleaner',
+                          module_path="ops.mapper.garble_characters_cleaner.process")

runtime/ops/mapper/garble_characters_cleaner/metadata.yml

@@ -0,0 +1,17 @@
+name: '文档乱码去除'
+name_en: 'Garbled Character Removal'
+description: '去除文档中的乱码和无意义的unicode。'
+description_en: 'Removes garbled characters and meaningless Unicode characters from
+  documents.'
+language: 'python'
+vendor: 'huawei'
+raw_id: 'GrableCharactersCleaner'
+version: '1.0.0'
+types:
+  - 'cleanse'
+modal: 'text'
+effect:
+  before: '文档乱码����'
+  after: '文档乱码'
+inputs: 'text'
+outputs: 'text'

runtime/ops/mapper/garble_characters_cleaner/process.py

@@ -0,0 +1,54 @@
+#!/user/bin/python
+# -*- coding: utf-8 -*-
+
+"""
+Description:
+    本插件实现将文档中乱码去除功能
+    实现逻辑：
+        1. 正则判断该字符的unicode编码是否在乱码范围内。若在范围内，则去除，不在范围内，则保留。
+        2. 运行前，加载乱码字符范围的配置文件，即charset.json。该json文件中，key为字符集名称，value为unicode编码范围的集合。
+
+Create: 2025/01/13
+"""
+import json
+import re
+import time
+from pathlib import Path
+from typing import Dict, Any
+
+from loguru import logger
+
+from datamate.core.base_op import Mapper
+
+
+class GrableCharactersCleaner(Mapper):
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self._file_path = str(Path(__file__).parent / 'resources' / 'charset.json')
+        self.unicode_grable_code_list = self.get_unicode_grable_code_list()  # 乱码unicode编码的十进制范围的集合
+        self.grable_re_compile = re.compile("[" + self.unicode_grable_code_list + "]")
+
+    def get_unicode_grable_code_list(self):
+        """获取乱码unicode编码范围"""
+        res = ""
+        with open(self._file_path, 'r', encoding='utf-8') as f:
+            charset_number_list = json.load(f)
+        for number_ranges in charset_number_list.values():
+            for number_range in number_ranges:
+                number_range_list = number_range.split(",")
+                if len(number_range_list) < 2:
+                    logger.error(f"number_range_list size is {len(number_range_list)}, formatting error")
+                    continue
+                res += number_range_list[0] + "-" + number_range_list[1]
+        return res
+
+    def execute(self, sample: Dict[str, Any]) -> Dict[str, Any]:
+        start = time.time()
+        sample[self.text_key] = self._grable_characters_filter(sample[self.text_key])
+        logger.info(
+            f"fileName: {sample[self.filename_key]}, method: GrableCharactersCleaner costs {time.time() - start:6f} s")
+        return sample
+
+    def _grable_characters_filter(self, input_data: str):
+        """去除文档中的乱码"""
+        return self.grable_re_compile.sub("", input_data)

runtime/ops/mapper/garble_characters_cleaner/resources/charset.json

@@ -0,0 +1,24 @@
+{
+  "注音符号东亚": [
+    "\u3100,\u312F"
+  ],
+  "拉丁文补充1": [
+    "\u00C0,\u00D6",
+    "\u00D8,\u00F6",
+    "\u00F8,\u00FF"
+  ],
+  "拉丁文扩展,A": [
+    "\u0100,\u017F"
+  ],
+  "拉丁文扩展,B": [
+    "\u0180,\u024F"
+  ],
+  "私人使用区域": [
+    "\uE000,\uF8FF",
+    "\\U000f0000,\\U000ffffd",
+    "\\U00100000,\\U0010fffd"
+  ],
+  "占位符": [
+    "\uFFFD,\uFFFD"
+  ]
+}